From mboxrd@z Thu Jan 1 00:00:00 1970 From: Claude Code Review Bot To: dri-devel-reviews@example.com Subject: Claude review: drm/gem: Consider GEM object reclaimable if shrinking fails Date: Tue, 31 Mar 2026 17:24:45 +1000 Message-ID: In-Reply-To: <20260330094848.2169422-2-boris.brezillon@collabora.com> References: <20260330094848.2169422-1-boris.brezillon@collabora.com> <20260330094848.2169422-2-boris.brezillon@collabora.com> X-Mailer: Claude Code Patch Reviewer Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit MIME-Version: 1.0 Patch Review Clean and correct. When `shrink()` returns false but the object hasn't been moved to a different LRU, it means the object is still reclaimable and `*remaining` should reflect that. The comment about the race being harmless (worst case: an extra scan) is adequate. The check `obj->lru == &still_in_lru` without holding the LRU lock is technically racy as noted, but the consequence is benign. Has two R-bs. No issues. --- Generated by Claude Code Patch Reviewer